Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: CM9000 tournaments (final standings - several different time controls)

Author: Graham Banks

Date: 21:10:47 05/25/04

Go up one level in this thread


On May 25, 2004 at 15:55:28, Kurt Utzinger wrote:

>On May 25, 2004 at 03:21:54, Graham Banks wrote:
>
>>40 moves in 10 minutes repeating (Athlon 2000)
>>
>>FINAL
>>Name                Wins    Draws   Losses  Games   Score
>>CM9000 Judge          26      32      12      70     42.0
>>CM9000 R1             25      30      15      70     40.0
>>CM9000 Mapi           23      29      18      70     37.5
>>CM Relentless         19      31      20      70     34.5
>>CM9000 Plague         17      33      20      70     33.5
>>CM9000 Cronus         12      40      18      70     32.0
>>CM 9000 Default       16      31      23      70     31.5
>>CM9000 SKR            15      28      27      70     26.0
>>
>>
>>40 moves in 20 minutes repeating (Athlon 3200+)
>>
>>Judge            38.5
>>SKR              38.5
>>Relentless       37.5
>>Cronus           36.5
>>R1               35.0
>>MAPI             33.0
>>Plague           32.0
>>Default          28.0
>>
>>
>>40 moves in 40 minutes repeating (Athlon 3200+)
>>
>>FINAL
>>Cronus        40.0
>>Judge         37.5
>>R1            37.0
>>SKR           37.0
>>Plague        34.5
>>Mapi          34.0
>>Relentless    32.5
>>Default       27.5
>>
>>
>>40 moves in 80 minutes repeating (Pentium 4)
>>
>>Name                Wins    Draws   Losses  Games   Score
>>CM9000_Plague         12      23       7      42     23.5
>>CM9000_Mapi           12      22       8      42     23.0
>>CM9000_Cronus         11      24       7      42     23.0
>>CM9000_R1             12      21       9      42     22.5
>>CM_Relentless          8      25       9      42     20.5
>>CM9000_Judge           9      21      12      42     19.5
>>CM9000_SKR             7      25      10      42     19.5
>>CM_9000_Default        5      23      14      42     16.5
>
>
>     Hi Graham
>     Did you notice one interesting thing: the difference
>     between first/last place is diminshing the longer
>     the time control chosen.
>     Kurt

Hi Kurt,
yes I had noticed this.
I believe that changing the material values slightly (especially the queen),
increasing the selectivity and leaving all positional values A/D and M/P values
at default might possibly produce the best results for CM9000. This is the tack
that I might try next.
I also have another idea I might try too.
I'll keep you posted.
Graham.



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.